Related Experiment Video
Updated: Mar 18, 2026
Modeling Neonatal Intraventricular Hemorrhage Through Intraventricular Injection of Hemoglobin
Published on: August 25, 2022
Noninvasive hemoglobin measurement in pediatric trauma patients
Mark Leo Ryan1, Angela C Maxwell, Lisa Manning
1From the Division of Pediatric Surgery, Department of Surgery (M.L.R., L.M., M.B-R., R.F.W.), and Department of Pediatric Emergency Medicine (A.C.M., J.D.J.), University of Tennessee Health Science Center, Le Bonheur Children's Hospital, Memphis, TN.
Introduction:
Hemorrhage is a major cause of preventable death secondary to traumatic injury. Diagnosis often requires multiple blood draws, which are psychologically stressful in pediatric patients. The Pronto device is a pulse co-oximeter that measures the total hemoglobin level using multiple wavelengths of light. The purpose of this study was to evaluate the accuracy of the noninvasive hemoglobin measurements relative to current invasive and point of care testing methods in pediatric trauma patients.
Methods:
We performed a prospective observational trial involving patients younger than 17 years presenting to a Level I pediatric trauma center. Following admission, blood was sampled from each patient for testing using an i-Stat device (point-of-care hemoglobin) and a complete blood count within our core laboratory (invasive hemoglobin). Noninvasive hemoglobin analysis was performed within 15 minutes of phlebotomy. Data were evaluated using Spearman correlation and Bland-Altman analysis.
Results:
Over 2 years, 114 patients had attempted noninvasive hemoglobin measurements, with a success rate of 89%. Mean ± SD age was 9.2 ± 5.1 years. Ninety percent of admissions were for blunt injury, 3% penetrating, 5% near drowning, and 1% burns. Mean invasive hemoglobin was 12.6 ± 1.9 g/dL, mean point-of-care hemoglobin was 12.2 ± 2.0 g/dL, and mean noninvasive hemoglobin was 12.3 ± 1.6 g/dL. Noninvasive hemoglobin values were strongly correlated with both invasive and point of care measurements (R = 0.672 and R = 0.645, respectively; p < 0.001). Bland-Altman analysis comparing noninvasive to point-of-care and invasive hemoglobin levels resulted in an estimated bias of -0.39 and -0.49, respectively.
Conclusion:
Noninvasive hemoglobin values had excellent correlation with both invasive and point-of-care hemoglobin measurements, although the device was not successful for all patients. Given the rapid availability of results and the lack of requirement of venipuncture, noninvasive hemoglobin monitoring may be a valuable adjunct in the initial evaluation and monitoring of pediatric trauma patients.
Level Of Evidence:
Diagnostic test study, level II.
